Then, the concept lattice was built by making use of the first N documents.Thereby, the concept matching was utilized to extract answer from the lattices.

• 接着利用这些文档中前N个文档构建概念格，进而采用概念匹配在格中抽取答案。